- Foundation
- About 16 min
- 5 practice questions
Experimental probability and relative frequency
Estimate probabilities from experiments and surveys using relative frequency, and use a sample to estimate how many times something happens in a whole population.
If something looks wrong, please tell us.
By the end of this lesson you can
- Find the experimental probability (relative frequency) of an event from the results of repeated trials.
- Check that the relative frequencies of all the outcomes add up to 1.
- Use data from a sample to estimate numbers in a larger population, and say what makes a sample good.
Will a paper cup tossed in the air land on its bottom, its top or its side? No formula tells us; the cup is not symmetrical. But we can toss it many times and count. Probability found this way, from evidence, is called experimental probability.
Trials, outcomes and relative frequency
Each repetition of an experiment is a trial, and each possible result is an outcome. The list of all possible outcomes is the sample space: for a coin it is {H, T}, for a die {1, 2, 3, 4, 5, 6}, for the cup {bottom, top, side}.
Experimental probability of an event = number of times the event happened ÷ total number of trials. This fraction is also called the relative frequency of the event.
Worked example 1: rolling a die
Suppose Aman rolls a die 60 times and gets a 4 exactly 12 times. Experimental probability of a 4 = 12/60 = 1/5 = 0.2, or 20%. The frequency of 4 is 12 (a count); the relative frequency is 0.2 (a fraction of all the trials).
Worked example 2: tossing a paper cup
Suppose a group tosses a paper cup 50 times and records: bottom 9, top 11, side 30. Find the experimental probability of each outcome.
| Outcome | Frequency | Relative frequency |
|---|---|---|
| Lands on its bottom | 9 | 9/50 = 0.18 |
| Lands on its top | 11 | 11/50 = 0.22 |
| Lands on its side | 30 | 30/50 = 0.6 |
| Total | 50 | 1 |
Working: divide each count by 50. Check: 0.18 + 0.22 + 0.6 = 1. Your own group's numbers will be different; that is what makes it an experiment.
Probability from survey data
Businesses, scientists and planners often estimate probabilities from data collected in the past or from a survey. Usually it is impossible to ask everybody, so they ask a sample and use it to estimate the whole population.
Estimating for a population from a sample
- Ask a clear question For example, 'How do you come to school?'
- Choose a fair sample Large enough, and drawn from all groups.
- Count the answers Make a frequency table.
- Find relative frequencies Count ÷ size of the sample.
- Scale up Relative frequency × size of the population.
Worked example 3: a school survey
Suppose 40 students, chosen from all classes of a school with 900 students, name their favourite game: cricket 18, football 10, kabaddi 8, chess 4. (a) What is the probability that a student chosen at random from the sample likes kabaddi best? (b) Estimate how many students in the school like kabaddi best.
Working: (a) 8/40 = 1/5 = 0.2. (b) Estimate = 0.2 × 900 = 180 students. This is an estimate, not an exact count: a different sample of 40 might give a slightly different answer.
What makes a good sample?
- Size: a larger sample usually gives a more dependable estimate than a small one.
- Representative: it should include all kinds of people in the population. Asking only the school cricket team about favourite games would give a biased answer.
- Honest recording: collect answers anonymously where people may feel shy, and do not leave out awkward results.
Key takeaways: experimental probability (relative frequency) = times the event happened ÷ number of trials. The relative frequencies of all outcomes add up to 1. Data from a fair sample gives an estimate for a whole population: multiply the relative frequency by the population size. Larger, representative samples give better estimates.
Check your understanding
-
1. A coin is tossed 80 times and shows heads 44 times. What is the experimental probability of heads?
- 0.44
- 0.5
- 0.36
- 0.55
Show answer
Answer: D. Experimental probability = 44 ÷ 80 = 0.55.
-
2. For a tossed cup, the experimental probability of landing on its bottom is 0.2 and on its top is 0.3. The only other outcome is its side. What is the experimental probability of the side?
- 0.5
- 0.3
- 0.25
- 0.6
Show answer
Answer: A. The relative frequencies of all outcomes add up to 1, so P(side) = 1 − 0.2 − 0.3 = 0.5.
-
3. In a survey of 50 students, 15 like football. What is the relative frequency of liking football?
- 15
- 0.3
- 0.15
- 0.5
Show answer
Answer: B. Relative frequency = 15 ÷ 50 = 0.3. The number 15 itself is the frequency, not the relative frequency.
-
4. A sample of 30 sweets from a jar of 600 has 9 green ones. About how many green sweets are in the jar?
- 90
- 270
- 180
- 9
Show answer
Answer: C. Relative frequency = 9 ÷ 30 = 0.3. Estimate = 0.3 × 600 = 180 green sweets.
-
5. A school wants to know students' favourite subject. Which sample is most suitable?
- The 20 students in the science club
- The 10 students who reach school first
- The toppers of each class
- 60 students chosen at random from all classes
Show answer
Answer: D. A good sample is large enough and represents all students. Choosing at random from all classes avoids the bias of a special group.
Sign in to keep track of the lessons you finish.